Exam 3

First thing, load the libraries that will be used.

The first code being used looks at data for college faculty salaries and total compensation from 1995, by Ranks, Tier, and State from 1995. The ranks are Assistant, Associate, and Full. The tier invovled the funding devoted to research vs the amount of funding for teaching; tier I spend mor on research than teaching and award PhD degrees.

Task I: Load and clean FacultySalaries_1995.csv file

Load and clean file:

df <- read.csv("./FacultySalaries_1995.csv")
names(df) <- make_clean_names(names(df))
names(df)

##  [1] "fed_id"                 "univ_name"              "state"                 
##  [4] "tier"                   "avg_full_prof_salary"   "avg_assoc_prof_salary" 
##  [7] "avg_assist_prof_salary" "avg_prof_salary_all"    "avg_full_prof_comp"    
## [10] "avg_assoc_prof_comp"    "avg_assist_prof_comp"   "avg_prof_comp_all"     
## [13] "num_full_profs"         "num_assoc_profs"        "num_assist_profs"      
## [16] "num_instructors"        "num_faculty_all"

df <- df %>%
  pivot_longer(c(avg_full_prof_salary,avg_assoc_prof_salary, avg_assist_prof_salary),
                names_prefix = "_salary", 
               values_to = "salary",
               names_to = "rank")
df <- df[!df$tier %in% "VIIB", ]

Creation of figure 1:

Task II: ANOVA salary summary

The ANOVA model tests the influence of “State”, “Tier”, and “Rank” on “Salary.”

aov.df <- aov(salary ~ tier + rank + state,  data = df)
summary(aov.df)

##               Df   Sum Sq Mean Sq F value Pr(>F)    
## tier           2  8119005 4059503 1116.49 <2e-16 ***
## rank           2 16521237 8260619 2271.92 <2e-16 ***
## state         50  4625868   92517   25.45 <2e-16 ***
## Residuals   3297 11987752    3636                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 128 observations deleted due to missingness

a <- summary(aov.df)
capture.output(a, file= "Salary_ANOVA_Summary.txt") 

Task III: Juniper Oils

This data is from a collaboration between Young Living Inc. and UVU Microbiology. A number of dead cedar trees were collected and the chemical composition of their essential oil content was measured. The hypothesis was that certain chemicals would degrade over time since they died in fires.

Clean data with chemical compounds:

df2 <- read.csv("./Juniper_Oils.csv")
names(df2)

##  [1] "SampleID"               "Project"                "Amplicon"              
##  [4] "Tree_Species"           "BurnYear"               "Latitude"              
##  [7] "Longitude"              "Field_Office"           "BLM_Fire_Name"         
## [10] "Tracking_Number"        "alpha.pinene"           "para.cymene"           
## [13] "alpha.terpineol"        "cedr.9.ene"             "alpha.cedrene"         
## [16] "beta.cedrene"           "cis.thujopsene"         "alpha.himachalene"     
## [19] "beta.chamigrene"        "cuparene"               "compound.1"            
## [22] "alpha.chamigrene"       "widdrol"                "cedrol"                
## [25] "beta.acorenol"          "alpha.acorenol"         "gamma.eudesmol"        
## [28] "beta.eudesmol"          "alpha.eudesmol"         "cedr.8.en.13.ol"       
## [31] "cedr.8.en.15.ol"        "compound.2"             "thujopsenal"           
## [34] "Yield_percent"          "Bolt_Surface_Area_cm2"  "Raw_Exit_Holes_per_cm2"
## [37] "Raw_Exit_Holes"         "Living_Larvae"          "ChemTotal"             
## [40] "ChemMean"               "YearsSinceBurn"

df2 <- df2 %>%
  pivot_longer(c("alpha.pinene","para.cymene","alpha.terpineol","cedr.9.ene","alpha.cedrene",
                 "beta.cedrene","cis.thujopsene","alpha.himachalene","beta.chamigrene",
                 "cuparene","compound.1","alpha.chamigrene","widdrol","cedrol","beta.acorenol",
                 "alpha.acorenol","gamma.eudesmol","beta.eudesmol","alpha.eudesmol",
                 "cedr.8.en.13.ol","cedr.8.en.15.ol","compound.2","thujopsenal"),
             values_to = "Concentration",
             names_to = "ChemicalID")

Task IV: Graph years since burn and concentration

Task V: Create Generalized Linear Model

Chemicals of significance (significant, as in P < 0.05) affected by “Years Since Burn.”

mod1 <- glm(data = df2, Concentration ~ YearsSinceBurn * ChemicalID)
#also still working on

Task V: Commit and push all code and files to GitHub

Terminal -> git add Exam 3 git commit -m “Exam 3” git push My folder for Exam 3 is not on GitHub.